Three postpartum outpatient treatment programs will be created and then compared for their effects on a sample of 144 cocaine-exposed black adolescent mothers and their infants. Based on a multi-level intervention program we previously used with adolescent mothers, the cocaine using mother-infant dyads will be randomly assigned at birth to one of three programs: 1) a drug rehab group (level 1) that receives 20 hours/week intensive drug rehab-for months including therapeutic and didactic drug rehab groups, 2 hours of NA and AA,, general education classes for high school equivalency test, vocational counseling classes, social support groups, and parenting and relaxation therapy classes; 2) a drug rehab daycare group (level 2) that receives the same drug rehab program plus infant care in a new infant nursery in a daycare program directed by the PI; and 3) a drug rehab plus infant daycare plus teacher-aide training group (level 3) that receives the same drug rehab program and infant daycare already discussed as well as simultaneous on-the-job parent/teacher aide training (20 hours/week) while caring for their own and others' infants in this infant nursery class. Cost advantages of this model are that the drug rehab program will meet the state requirements for financial aid and the free teacher aides will help us meet teacher child ratios for a relatively low cost daycare program. Finally, all three groups will receive a 6-month follow-up rehab program (7 hour /week) including one hour each of drug rehab, NA and AA meetings, educational/vocational, social support, parenting and relaxation therapy classes. The groups will be evaluated at 1, 3, 6( 12, 18, 24, and 36 months. Assessments will include maternal interviews (drug use, lifestyle, mental health), random urine drug checks, infant developmental exams (Brazelton, Fagan, MAI Bayley, Binet) physical exams (health and growth), neurological evaluations and mother-infant interactions. School-age follow-up with the infants (and their mothers) will be enabled by a school system risk registry. Group comparison, regression and cost benefit analyses will be performed on the data.